Developing of the evaluation of water inrush risk from coal seam floor
YAO Hui;YIN Huichao;YIN Shangxian;HOU Enke;BI Meng;LIAN Huiqing;XIA Xiangxue;LIANG Manyu
西安科技大学 地质与环境学院华北科技学院 河北省矿井灾害防治重点实验室中国矿业大学(北京)地球科学与测绘工程学院防灾科技学院 信息工程学院华北科技学院 安全工程学院
回顾了煤层底板突水危险性评价的发展历程,提出其危险性评价的体系:指标(指标体系的建立)—方法(评价方法的选取)—工具(处理工具的革新),并对3个环节进行总结。指出指标体系的发展不再是因素集的扩充,而是因素与因素之间非线性关系的处理以及对以开采条件及地质条件为两大基本要素集的化繁为简;将现有方法依据处理数据的逻辑分为3类:以数据的基础信息为基准,将原始数据对评价对象所产生的大小、高低、优劣性影响进行考量、排序及综合形成评价结果的第一类方法;对数据列进行人工评判、加工分析、拓展和延伸,挖掘数据的潜在信息,并形成最终评价结果的第2类方法;整理具有相同指标的数据集,通过数据信息处理技术发现数据间的共有信息,从而获得最终评价结果的第3类方法。指出未来评价方法的发展方向一方面是对突水系数法的传承,修正其在厚、巨厚、极薄隔水层的不良表现,另一方面是对机器学习新型方法的创新,对其本身及组合模型进行开发与应用。 提出了处理工具所需实现的三大目标:矿井立体化模型的建立、评价结果的动态化演示、“定位、定量、定概率”三定指标的实现。分别探讨了三者面临的问题并阐述具体解决手段。在上述基础上,总体阐明了煤层底板突水危险性评价体系各环节的研究展望。
Reviews the development process of the evaluation of water inrush risk from coal seam floor. Then the system of risk evaluation is put forward: indexes (establishment of index system), methods (selection of evaluation methods) and tools (innovation of handling tools), and the three steps are summarized. The study indicates that the development of index system is no longer the expansion of factor sets, but the treatment of non-linear relationship among factors, as well as the simplification of the two basic factor sets: mining and geological conditions. The existing methods are divided into three categories according to the logic of data processing. Based on the basic information of the data, the first type is to consider, sort, and synthesize three effects of the original data on the evaluated object: size, height, advantages and disadvantages, thus forming the evaluation result. The second type includes assessment, analysis, expansion and extension of the data, then the potential information is discovered to form the final result. The third type is to organize data sets with the same indexes, and find common information among data through related processing technology to obtain the result. The development direction of future evaluation methods covers two aspects. On the one hand, it aims to inherit the water inrush coefficient method and improve its poor performance in thick, extremely thick, and extremely thin water-resisting layers. On the other hand, it aims to innovate new methods of machine learning, then develop and apply them and their combined models. Besides,three goals that the processing tool needs to achieve are proposed:build a three-dimensional model of a mine,the realization of dynamic demonstration,positioning,quantitation and probability.The problems faced by the three parts are discussed and specific solutions are elaborated.On the basis of the above,the research prospect of all links of water inrush risk evoluation system from coal seam is globally clarified.
mining under pressure;water inrush from coal seam floor;risk evaluation;water inrush coefficient;neural network
主办单位:煤炭科学研究总院有限公司 中国煤炭学会学术期刊工作委员会